1. Field of the Invention
The present invention relates generally to the fields of immunology and tumor therapy. More particularly, it concerns homoconjugates of antibodies which arrest cell growth and/or signal apoptosis in the tumor cells. The components of these homoconjugate antibodies can comprise a wide variety of antibodies but often and surprisingly do not require an Fc region to function and importantly activate fewer, or even no, undesired immunological reactions.
2. Description of Related Art
During the past two decades, a variety of monoclonal antibodies (MAbs) have been selected for clinical use based on their effector functions and the most encouraging results have emerged from the treatment of NHL (Hjekman et al., 1991; Press et al., 1987; Meeker et al., 1985; Waldmann, 1992; Hale et al, 1988; Dyer et al., 1989; Hamblin et al., 1987; Brown et al., 1989; Rankin et al., 1985) and, in particular, when radioimmunoconjugates were used (Kaminski et al., 1993; Press et al., 1993). Two examples are the epithelial cell-reactive MAb, 17.1A (Herlyn and Koprowski, 1982; Riethmxc3xcller et al., 1994) and the lymphoma reactive MAb, CAMPATH-1 (Dyer et al., 1989; Hale et al., 1988). In this regard, there is considerable experimental (Hooijberg et al., 1995a; Hooijberg et al., 1995b), and some clinical (Riethmxc3xcller et al., 1994; Hale et al., 1988) evidence to indicate that effector functions play an important role in the anti-tumor activity of MAbs. Those MAbs which do not fix complement or mediate antibody-dependent cell mediated cytotoxicity (ADCC) have been converted into useful ones by chimerization (Morrison and Oi, 1989; Adair, 1992), by generating switch variants (Hale et al., 1985; Kaminski et al., 1986; Denkers et al., 1995) or by preparing cytotoxic immunoconjugates (Dillman, 1994; Ghetie et al., 1996b; Hellstrxc3x6m et al., 1995).
It has been demonstrated that the crosslinking of mIg on human Daudi cells initiates a cascade of signals leading to the induction of both apoptosis and CCA (Racila et al., 1996). Using antisense oligonucleotides, it was shown that the mig-associated Lyn tyrosine kinase was required for anti-Ig mediated CCA but not for the induction of apoptosis (Racila et al., 1995). It was also shown that Lyn was critical for the induction of CA by anti-CD19 (Racila et al., 1995). These results suggest that the signaling pathways leading to CCA and apoptosis might bifurcate at an early stage of BCR crosslinking.
In attempting to further distinguish the different signaling pathways, studies were conducted to understand signal transduction initiated by crosslinking BCRs. Recent evidence supports a mechanism whereby TCR-induced apoptosis is dependent on Fas/Fas ligand interactions. Thus, crosslinking TCRs results in transient upregulation of the Fas ligand in T-lymphoma cells (Dhein et al., 1995; Brunner et al., 1995; Ju et al., 1995). Apoptosis induced by T cell receptor/T cell receptor (TCR) crosslinking is markedly inhibited by either anti-Fas F(abxe2x80x2)2 fragments (which are not cytotoxic) or soluble Fas-Fc (Dhein et al., 1995; Brunner et al., 1995; Ju et al., 1995). These results indicate that in T-lymphoma cells, apoptosis induced by TCR activation results from the induction of the expression of the Fas ligand and its interaction with Fas resulting in the activation of the Fas signaling pathway.
In contrast to T cells, crosslinking of membrane IgM on Daudi cells (which constitutively express Fas) did not induced synthesis of Fas ligand (Racila et al., 1996). Furthermore, a noncytotoxic fragment of anti-Fas that blocked T cell receptor-mediated apoptosis did not block anti-xcexc induced apoptosis. Hence, in the B lymphoma cells, Daudi, apoptosis induced by signaling via IgM is not mediated by the Fas ligand. Similar results were obtained using a murine lymphoma (Scott et al., 1996). More recent studies suggest that the signaling is related to TNFxcex1 and TNFxcex2 which are the only members of the TNF family which are upregulated in Daudi cells after crosslinking IgM. These studies suggest that IgM and CD19 may utilize different signaling pathways and that the activation pathways in Daudi cells are different from those induced via TCRs in T lymphoma cells.
Hence, like anti-CD40 (Marches et al., 1996), anti-CD19 enhanced the apoptosis induced by anti-xcexc. This finding is relevant for clinical strategies because it is obviously more desirable to kill a tumor cell than to induce CCA. The combination of anti-xcexc and anti-CD19 was studied in SCID/Daudi mice (where there is no serum IgM). Using multiple titrations of the two agents in the animals with a relatively long read-out period of 150 days. (Anti-xcexc will not be useful as a therapeutic agent in normal mice and humans, but is useful as a demonstration model.) In studying several MAbs which react with these molecules, it was found that none signaled as well as the anti-CD19 (HD37). It is postulated that either the target molecules could not readily mediate signals when bound to MAbs and/or that the ability of the MAbs to crosslink their target molecules was poor.
The therapeutic potential of MAbs has not been fully realized because there has been no paradigm for predicting which properties of the MAb are essential and how MAbs interact with other therapeutic modalities. For example, until recently, it was believed that the antitumor activity of a MAb was due solely to its conventional immunological effector mechanisms (i.e., ADCC, Cxe2x80x2 fixation) (Dyer et al., 1989; Hale et al., 1984). Although this is true in certain instances, there is accumulating evidence that the antitumor activity of many MAbs is due to their ability to signal growth arrest to death (Trauth et al., 1989; Yefenof et al., 1993) or to their ability to inhibit cell traffic (Zahalka et al., 1993), cell-cell interactions (Zahalka et al., 1995; 25) or extravasation (Ruiz et al., 1993; Akiyama et al., 1995).
Recently it has been shown that MAbs can exert anti-tumor activity in other ways, e.g. by inhibiting metastases (Qi et al., 1995), tumor cell-substrata interactions (Guo et al., 1994), or tumor cell extravasation (Edward, 1995). In addition, it has been reported, that some MAbs can signal growth arrest and/or apoptosis of tumor cells, by acting as agonists (xe2x80x9cnegative signalingxe2x80x9d) (Ghetie et al.. 1992; Ghetie et al., 1994; Vitetta and Uhr, 1994; Trauth et al., 1989; Page and Defranco, 1988; Bridges et al., 1987; Funakoshi et al., 1994; Beckwith et al., 1991; Schreiber et al., 1992; Scott et al., 1985). xe2x80x9cNegative signalingxe2x80x9d is herein defined as the inhibition of cell growth by cell cycle arrest or the induction of apoptosis (programmed cell death). Indeed, in the case of B cell lymphoma, there is compelling evidence that both anti-idiotype (Levy and Miller, 1990; Hamblin et al., 1980) and anti-CD19 MAbs (Ghetie et al., 1992; Ghetie et al., 1994) exert their anti-tumor activities predominantly, if not exclusively, by signaling growth arrest and apoptosis. Other MAbs which also have signaling properties include anti-Fas (Trauth et al., 1989), anti-CD40 (Funakoshi et al., 1994), anti-Class II MHC (Bridges et al., 1987), anti-Her-2 (Scott et al., 1991), anti-Ley (Schreiber et al., 1992) and anti-IgM (Vitetta and Uhr, 1994; Page and Defranco, 1988; Beckwith et al., 1991; Scott et al., 1985). Furthermore, negative signaling can sometimes be optimized by hypercrosslinking with secondary antibodies or by using xe2x80x9ccocktailsxe2x80x9d of primary antibodies (Marches et al., 1996).
In the case of anti-CD 19, only a small percentage of MAbs can deliver growth inhibiting signals to neoplastic B cells and these require the addition of very large (i.e. hypersaturating) concentrations of antibody (Ghetie et al., 1994). Unfortunately, these large concentrations of antibody can activate unwanted immune responses which clear the therapeutic antibodies from the system and reduce the efficacy of treatment. Clearly, it is desirable to eliminate these unwanted immune responses in order to increase the efficacy of treatment, but to date, the means to eliminate these responses is not available.
The therapeutic or diagnostic usefulness of a monoclonal antibody (MAb) is dependent upon several factors. The MAb must possess sufficient binding affinity and a relatively high avidity for an antigen. The avidity of a MAb is based on the valency of the antibody (and the antigen) and the quaternary arrangement of the interacting components. The physical size of the molecule is also an important limiting factor. Thus while hundreds of MAbs recognize tumor antigens, few MAbs have proven useful for the diagnosis and treatment of tumors and neoplastic diseases, because they are not specific enough, they fail to have sufficient affinity or avidity or are too large to reach their antigens.
Part of the difficulty in designing antibody conjugates which possess sufficient affinity and avidity for tumor antigens and are of an appropriate isotype or subclass to efficiently initiate cell cycle arrest or apoptosis is that few MAbs possess sufficient affinity and avidity to be useful. For example, IgG molecules are monomers and have low valency; and thus, they have low avidity. IgM molecules have a higher valency but their size limits their ability to penetrate tissue and reach the desired antigenic target.
Homoconjugates of IgG molecules have been designed which possess increased valency and thus have higher avidity and are better able to promote effector function (PCT application WO92/04053; Wolff et al., 1992; 1993; Caron et al., 1992). But these homoconjugates possess two or more Fc regions; and thus, they can still elicit undesired immune responses which reduce successful treatment of a tumor or neoplastic disease. And although increasing their valency results in increased avidity, the physical sizes of the homoconjugates are also increased which may limit or even prevent them from being able to physically reach target epitopes.
There is a clear and present need for therapeutic MAbs with enhanced avidity which do not produce unwanted immune responses and are relatively small molecules which are capable of wide biodistribution. The present invention provides such compositions as conjugates of MAbs which do not possess an Fc region and yet are surprisingly active at signaling cell cycle growth arrest and/or apoptosis. The compositions of the present invention provide the further improvements of being relatively smaller than conventional conjugates of IgG MAbs, and are thus capable of wider biodistribution, and producing fewer unwanted immune responses because they do not signal effector functions.
The present invention provides conjugates of monoclonal antibodies (MAbs) which can lack Fc regions and yet are as effective, or possibly more effective, at signaling cell cycle arrest and/or apoptosis of tumor cells than comparable MAb conjugates which contain Fc regions. Surprisingly, conjugates of the present invention can be comprised of MAbs which in their monomeric or unconjugated form have little or substantially no anti-tumor activity, indicating that the binding to and crosslinking of cell surface antigens to the conjugates elicits a negative signal. Conjugates of the present invention are relatively smaller and capable of better biodistribution and, because they cannot recognize Fc receptors and stimulate effector function, these novel conjugates do not elicit unwanted immune responses.
The invention provides conjugates of two or more monoclonal antibodies, wherein the conjugates comprise at least one monoclonal antibody that does not comprise an Fc region. Conjugates comprise two, three, four or more monoclonal antibodies. In certain embodiments, no monoclonal antibody, comprised in any conjugate, comprises an Fc region. In further embodiments the conjugate comprises a monoclonal antibody that comprises an Fab or an Fv region or a fragment thereof.
The invention further provides a conjugate that comprises a monoclonal antibody that asserts substantially no anti-neoplastic activity, or even no anti-neoplastic activity, in an unconjugated form. The anti-neoplastic activity is an anti-proliferative activity and preferably is expressed as signaling growth arrest or apoptosis (i.e., programmed cell death, PCD).
In one aspect of the invention the conjugate comprises a monoclonal antibody that is a tumor reactive monoclonal antibody. In one particular illustration of the invention, preferred conjugates comprise a monoclonal antibody that is an IgG, IgA, IgD or IgM monomer and particularly preferred monomers are mammalian monomers, such as murine or human monoclonal antibody monomers. Exemplary monoclonal antibodies include, but are not limited to, an anti-CD19, anti-CD20, anti-CD21, anti-CD22, anti-HER2, (for example, HER66, HER50 and HER164) Mabs which react with breast tumors, ovarian tumors, prostate tumors, and/or lung tumors.
In another aspect of the invention the conjugates comprise antibodies that are conjugated via hypercrosslinking. In further aspects, the antibodies are conjugated via one or more covalent bonds such as disulfide bonds, thioether bonds or other covalent bonds which may function in vivo. These can be genetically engineered bonds as well as chemically engineered or even a combination of genetically engineered recombinant bonds and chemical bonds.
In other embodiments of the invention, the conjugates also comprise other proteins which may also be dimerized either by recombinant, chemical or recombinant and chemical means.
In a preferred embodiment the conjugate has a valence of at least three. In a more preferred embodiment the conjugate has a valence of at least four and sometimes five.
In a particular embodiment of the invention the conjugate is a homoconjugate and in another embodiment the conjugate is a heteroconjugate. As used herein, xe2x80x9chomoconjugatexe2x80x9d refers to a conjugate comprised of a single species of monoclonal antibody and xe2x80x9cheteroconjugatexe2x80x9d refers to a conjugate comprising two or more different species of monoclonal antibodies.
The invention further provides a method of making a conjugate of two or more monoclonal antibodies that comprises obtaining a first monoclonal antibody that does not comprise an Fc region; obtaining a second monoclonal antibody; and conjugating the first monoclonal antibody to the second monoclonal antibody.
In a particular embodiment of the invention the first monoclonal antibody is a monoclonal antibody that asserts anti-neoplastic activity in a conjugated form. In a further embodiment of the invention the second monoclonal antibody is a monoclonal antibody that asserts anti-neoplastic activity in a conjugated form. In certain embodiments both the first monoclonal antibody and the second monoclonal antibody are monoclonal antibodies that assert anti-neoplastic activity in a conjugated form.
In one aspect of the invention the first monoclonal antibody is a monoclonal antibody that asserts substantially no anti-neoplastic activity in an unconjugated or monomeric form. In another aspect of the invention the second monoclonal antibody is a monoclonal antibody that asserts substantially no anti-neoplastic activity in an unconjugated or monomeric form. In still another aspect of the invention both the first monoclonal antibody and the second monoclonal antibody are monoclonal antibodies that assert substantially no anti-neoplastic activity in an unconjugated or monomeric form.
In yet another embodiment is provided a method further consisting of obtaining a third, or even a fourth, monoclonal antibody; and conjugating the third, or even fourth, monoclonal antibody to the conjugate.
The invention provides a method of signaling an anti-neoplastic activity that comprises obtaining a conjugate of two or more monoclonal antibodies, wherein the conjugate comprises a monoclonal antibody that does not comprise an Fc region and wherein the conjugate comprises a monoclonal antibody that asserts anti-neoplastic activity in a conjugated form; and contacting a neoplastic cell or tumor with the conjugate.
The invention further provides a method of detecting the presence of a neoplastic disease that comprises contacting a biological sample, that is suspected of comprising a neoplastic antigen, with a conjugate that comprises a monoclonal antibody and screening for an immunological reaction.
In an additional aspect the invention provides a pharmaceutical composition comprising a pharmaceutically acceptable carrier and a conjugate comprising a monoclonal antibody.
An xe2x80x9cFc regionxe2x80x9d is herein defined as that region of an immunoglobin (Ig) composed of the C-terminal half of the heavy chains including that part of the hinge region containing the inter-heavy chain disulphide bridges; which may be released enzymatically from the Ig molecule; and specifically recognizes and binds to an Fc receptor such that effector function, such as phagocytosis or inflammatory response, is expressed.
Thus it will be readily understood that, herein, a monoclonal antibody that lacks an Fc region comprises an antibody which does not stimulate effector function by binding to an Fc receptor. The hinge regions of Igs vary considerably in length; thus, it is contemplated that a monoclonal antibody lacking an Fc region may contain a fragment of Fc region of almost any intermediate length such that effector function is not expressed in the presence of an Fc receptor. Alternatively, mutations or alterations to an Fc region may be introduced such that the length of the Fc fragment is about equivalent to the native Fc region but the altered Fc fragment cannot stimulate effector function in the presence of an Fc receptor. In certain embodiments, the preferred length of hinge region is preferably being limited by the ease of preparation and use in the intended conjugation protocol.
It will be readily understood that xe2x80x9cintermediate lengthsxe2x80x9d, in these contexts, means any length between the quoted ranges, such as 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, etc.; 50, 51, 52, 53, etc.; 70, 71, 72, 73, 74, 75, etc.; 100, 101, 102, 103, etc.; 120, 121, 122, etc.; 140, 141, 142, etc.; including all integers through the 1-150 ranges, and the like.
The antibodies used to prepare the conjugates of the present invention may be obtained by immunizing an animal with a mammalian, preferably a murine, or even more preferably a human protein or peptide that comprises a tumor antigen, and collecting the resultant antibodies. The protein or peptide comprising the tumor antigen may be prepared by obtaining a sample comprising the tumor antigen from any national or international registry which maintains tumor and cancer cell lines, such as the National Cancer Institute Tumor Registry, and using methods well-known to those of skill in the art. The Mabs can be prepared by phage display, cloning of cDNAs or any other molecular biological techniques.
What is meant by a monoclonal antibody having xe2x80x9csubstantially no anti-neoplastic activityxe2x80x9d is that any activity that is detected, as measured by 3H-thymidine inhibition as described herein, is not statistically significant (as determined, for example, by the Student t test) when compared to a control. What is meant by a monoclonal antibody having xe2x80x9cno anti-neoplastic activityxe2x80x9d is that no activity is detected by 3H-thymidine inhibition, as described herein.
It will be understood that by xe2x80x9cneoplasticxe2x80x9d is meant a tumorous condition which may comprise diffuse or well-differentiated tumor cells and affect reproductive tissue, such as ovaries, breast, testes, neural tissue, the alimentary tract, lymph tissue, bone marrow, lung, prostate, liver or any other type of tissue subject to neoplastic diseases.
Those of skill in the art will recognize that a wide variety of epitopes may be expressed by neoplastic or cancer cells as discussed by DeVita et al. In Cancer: Principles and Practice of Oncology, Fourth edition, Lipincott, Philadelphia, 1993. It will be further understood that any monoclonal antibody that recognizes such epitopes are useful in the present invention. A listing of many of these monoclonal antibodies may be found in the American Type Culture Collection: Catalogues of Animal Viruses and Antisera, Chlamydiae and Rickettsiae (Rockville, Md., USA), incorporated herein by reference. Additional monoclonal antibodies may be isolated by following the protocols described herein and using a tissue or cell line obtained from any national or international tumor registry, such as the National Cancer Institute Tumor Registry.
Other objects, features and advantages of the present invention will become apparent from the following detailed description. It should be understood, however, that the detailed description and the specific examples, while indicating preferred embodiments of the invention, are given by way of illustration only, since various changes and modifications within the spirit and scope of the invention will become apparent to those skilled in the art from this detailed description.